J . A m . Chem. SOC.1987, 109, 5538-5539
5538
proteolysis are not due to severely different substrate binding modes but instead to the association of the scissile carbonyl with either zinc or Arg-127, respectively. In addition, there may also be different rate-determining steps4 or an altogether different mechanism for sterically hindered substrates.6 The synthetic reaction catalyzed by CPA'o,llrecently has been exploited toward the observation of chemical interrnediates.l2 The binding mode observed in the ternary complex reported is one which may be mistaken for such an intermediate, and this mode may occur under conditions of excess substrate or product. This binding mode may also occur under conditions of low temperature,20 where product diffusion may be hindered due to viscosity of the cryobuffer. Nevertheless, such a binding mode could easily confound spectroscopic attempts to identify catalytic intermediates involving a ligand-saturated metal ion.21 (20) Kuo, L. C.; Makinen, M. W. J . Am. Chem. SOC. 1985, 107, 5255-5261. (21) We thank the National Institutes of Health for Grant GM-06920 in support of this research and also W. R. Grace & Co. for their financial support. Additonally, D.W.C. thanks AT&T Bell Laboratories for a doctoral fellowship.
Selective 0-Atom Transfer from Nitrous Oxide to Hydride and Aryl Ligands of Bis(pentamethylcyc1opentadienyl)hafnium Derivatives G. Alan Vaughan, Peter B. Rupert, and Gregory L. Hillhouse*
Searle Chemistry Laboratory, Department of Chemistry The University of Chicago Chicago, Illinois 60637 Received April 29, 1987 One of today's greatest chemical challenges is the selective oxidation of organic substrates.' Reagents energetically capable of effecting useful oxidations are often so reactive that indiscriminate attack of the substrate results; others produce byproducts that subsequently become involved in undesirable side reactions. In this light, an attractive oxidant for oxygen-transfer reactions is nitrous oxide, N=N=O, a molecule unstable with respect to its constituent elements (AGfo = 25 kcal/mol) and therefore a thermodynamically potent oxidant, yet one possessing impressive kinetic inertness, reflected in a 59-kcal/mol activation barrier for its thermal decomposition (a unimolecular, spin-forbidden process).2 Importantly, 0-atom-transfer reactions involving N 2 0 could be very clean since the sole byproduct is N2. With very few exception^,^ when N 2 0 reacts with transition-metal complexes, nitrogen is extruded and oxo complexes are formed (eq l).4 It M" N 2 0 O=Mn+2 + N 2 (1)
+
-
(1) Sheldon, R. A,; Kochi, J. K . Metal-Catalyzed Oxidations of Organic Compounds; Academic: New York, 1981. (2) Jolly, W. L. The Inorganic Chemistry of Nitrogen; W . A. Benjamin: New York, 1964; pp 70-71. (3) (a) Armor, J. N.; Taube, H. J . Am. Chem. SOC.1969, 91, 6874. (b) Armor, J. N.; Taube, H. Ibid. 1971, 93, 6476. (c) Pu, L. S.; Yamamoto, A.; Ikeda, S . Chem. Commun. 1969, 189. (d) Yamamoto, A.; Kitazume, S.;Pu, L. S.; Ikeda, S. J . Am. Chem. Sor. 1971, 93, 371. (e) Schrauzer, G. N.; Kiefer, G. W.; Doemeny, P. A,; Kisch, H. Ibid. 1973, 95, 5582. (f) Sellmann, D.; Kleinschmidt, E. 2. Naturforsch., B: Anorg. Chem., Org. Chem. 1977, 328, 795. (9) Cooper, J. N.; Green, M. L. H.; Couldwell, C.; Prout, K. J . Chem. SOC.,Chem. Commun. 1977, 145. (h) Banks, R. G. S . ; Henderson, R. J.; Pratt, J. M. J . Chem. SOC.A 1968, 2886. (4) (a) Bottomley, F.; Brintzinger, H . J . Chem. SOC.,Chem. Commun. 1978, 234. (b) Bottomley, F.; White, P. S. Ibid. 1981, 28. (c) Bottomley, F.; Lin, I. J. B.; White, P. S. J . Am. Chem. SOC.1981, 103, 703. (d) Bottomley, F.; Lin, I. J. B.; Mukaida, M. Ibid. 1980, 102, 5238. (e) Bottomley, F.; Paez, D. E.; White, P. S.Ibid. 1981, 103, 5581. ( f ) Bottomley, F.; Paez, D. E.; White, P. S . Ibid. 1982, 104, 5651. (9) Liu, H.-F.; Liu, R.-S.; Liew, K. Y.; Johnson, R. E.; Lunsford, J. H. Ibid. 1984, 106, 4117. (h) Many reports describe chemisorption and catalytic decomposition of N 2 0 on metal surfaces. For a recent example, see: Tan, S.A,; Grant, R. B.; Lambert, R. M. J . Catal. 1987, 104, 156.
0002-7863/87/1509-5538$01.50/0
Scheme I
3
0
\
"$
5
N
.Nz
-
,ai Cp',Hf_
Ph
"
";
A
was our thought that, since the transformation depicted in eq 1 results in an increase in the formal oxidation state of the metal by +2, interesting reactivity between N 2 0 and coordinated ligands might be observed in do systems where oxidation at the metal center is precluded. It has been previously shown that organoazides (N=N=NR)' and diazoalkanes (N=N=CR2),6 molecules isoelectronic with nitrous oxide, react cleanly with many Group 4 do metallocene derivatives; we now report that N,O can be used under mild conditions to selectively oxidize hydrido (to hydroxy) and aryl (to aryloxy) ligands in these systems. Nitrous oxide reacts at -78 OC with toluene solutions of Cp*,HfH2 (1)7 (Cp* T-C,Me,) in a 1:l stoichiometry to afford N2 and Cp*,HfH(OH) (2)8 quantitatively (eq 2).9 Interestingly,
-78 DC
Cp*,HfH, + N=N=O Cp*2HfH(OH) + N2 (2) 1 2 no intermediates were detected ('H N M R ) for this 1 2 conversion, in contrast to the analogous reactions of arylazides with 1 that yield isolable insertion products that subsequently lose N2 to afford amido species (eq 3).5a Diazoalkanes likewise react with 1 but give stable insertion products that are not prone to N2 extrusion (eq 4).6c Kinetic measurements (VT 'H NMR) show 1
+ RN,
0
1
oc
-
80 'C
Cp*,HfH(NHNNR) Cp*2HfH(NHR)
+ R2CN2
-+
Cp*2HM(v2-NHN=CR2)
+ N, (3) (4)
that the reaction in eq 2 is a second-order process with d[2]/dt 0: [1][N20];an Eyring plot using data obtained under pseudofirst-order conditions ( [N20] >> [ 11) yielded activation parameters of AH* = 11.7 h 0.5 kcal/mol and A S * = -12 i= 2 eu.9310 (l-d2) , Comparison of these data with measurements on C P * ~ H ~ D indicated a modest kinetic deuterium isotope effect (at -70 OC, k H / k D= 1.6 (1)) for the reaction. With an excess of N 2 0 at higher temperatures (80 "C), 2 reacts further to give a complex mixture containing C P * , H ~ ( O H )(-40%). ~ Treatment of solutions of Cp*,HfH(Ph) (3)7with N 2 0 (80 OC, 2 h) results in N2 evolution and competitive oxidation of the hydride and phenyl ligands of 3, cleanly yielding c ~ * ~ H f ( o H ) ( P h ) (4) and Cp*,HfH(OPh) (5) (with no spectroscopically observed intermediates) in a molar ratio of 4:5 = 3:2 (eq 5)." At higher ( 5 ) (a) Hillhouse, G. L.; Bercaw, J. E. Organometallics 1982, 1, 1025. (b) Chiu, K. W.; Wilkinson, G.; Thornton-Pett, M.; Hursthouse, M. B. Polyhedron 1984, 3, 79. (6) (a) Gambarotta, S.; Basso-Bert, M.; Floriani, C.; Guastini, C. J . Chem. SOC.,Chem. Commun. 1982, 374. (b) Gambarotta, S.; Floriani, C.; ChiesiVilla, A.; Guastini, C. Inorg. Chem. 1983, 22, 2029. (c) Moore, E. J. Ph.D. Thesis, California Institute of Technology, Pasadena, CA, 1984. (7) Roddick, D. M.; Fryzuk, M. D.; Seidler, P. F.; Hillhouse, G. L.; Bercaw, J. E. Organometallics 1985, 4, 97. (8) Alternatively, 2 can be prepared from 1 and water. Hillhouse, G. L.; Bercaw, J. E. J . Am. Chem. SOC.1984, 106, 5472. (9) Experimental procedures, spectral data, elemental analyses, and kinetic data are given in the Supplementary Material. (10) The rate was shown to have an approximate first-order dependence on [N,O], but an exact determination of this was prevented because of the difficulty in quantifying the concentration of nitrous oxide in solution.
0 1987 American Chemical Society
5539
J . A m . Chem. SOC.1987, 109, 5539-5541 temperatures (100 "C) or longer reaction times (12 h) 5 reacts with N 2 0 to give N 2 and the mixed hydroxyphenoxy complex C P * ~ H ~ ( O H ) ( O (6),11 P ~ ) shown in eq 6, but 4 does not react with Cp*,HfH(Ph) -l
+ N2O
J
80
-0.6Cp*zHf(OH)(Ph) 4
+ N20
O C
+
-0.4Cp*ZHfH(OPh) 5
+ N2
(5)
100 O C
Cp*zHf(OH)(OPh) + NZ (6) 6 N 2 0 below its decomposition temperature (140 "C)." The oxidation of the phenyl and hydride ligands of 3 at comparable rates argues against an insertion mechanism for N 2 0 activation analogous to that seen for RN, and R2CN2(eq 3 and 4). While the intermediacy of [Cp*,Hf(ONNH)(Ph)] in the formation of 4 is plausible, [Cp*IHfH(ONNPh)] is not a reasonable precursor to 5 because of the high activation barrier expected for the required phenyl migration from nitrogen (owing to the multiple bond character inherent in such N-Ph linkages).', A more likely mechanism consistent with these data involves initial O-coordination of N 2 0 to the coordinatively unsaturated Hf center followed by rate-determining hydride (or aryl) migration with concomitant loss of N 2 (Scheme I).I4 The apparent differing modes of interaction for these isoelectronic molecules ( N 2 0vs. RN,, R2CN2) with Group 4 metallocene derivatives may well be attributable to steric factors. Since oxygen-transfer reactions using Group 4, do metals with the common oxidants O2Isand t-BuOOH (TBHP)l6s1' comprise an important, well-studied class,'* it is worth noting an interesting feature of these reactions (that contrasts with ours using N 2 0 ) : they proceed via intermediate alkylperoxy ligands and, often, deriued radical^.^^^^^ Furthermore, the thermal generation of radicals in these systems can limit their synthetic utility: TBHP converts many Cp*,HfR2 complexes to the corresponding Cp*2Hf(OR)(O-t-Bu)but is ineffectual in oxidizing the aryl ligand of 3 due to 0-0 homolysis of the isolable Cp*,Hf(OO-t-Bu)(Ph) intermediate to give 4 (compare this with eq 5).16" In summary, these findings are significant because they demonstrate a fundamentally new mode of nitrous oxide reactivity with transition-metal complexes: direct 0 - a t o m transfer from N 2 0 to metal-ligand bonds instead of the commonly observed 0-abstraction to form metal-oxo specie^.^ We are currently 5
~~~~
-
-
-
+ H 2 0 4; 1 + HOPh 6. See ref 9 for details. (12) Thermal decomposition with loss of benzene occurs on heating 4 to 140 'C. (13) ,(a) Pauling, L. The Nature of the Chemical Bond, 3rd ed.; Cornell University: Ithaca, NY, 1960; pp 280, 296. (b) Streitweiser, A., Jr.; Heathcock, C. H . Introducrion to Organic Chemistry, 3rd ed.; Macmillan: New York, 1985; p 691ff. (14) As pointed out by a referee, there is no precedent for 0-coordinated N,O, but it is reasonable that highly oxophilic early-metal do centers would promote this binding m ~ d e . ' ~Nevertheless, ~.~ initial N-coordination and rearrangement or concerted rearrangement/insertion should be considered as a mechanistic possibility. Attempts to measure an isotope effect using Cp*2Hf(D)(Ph)resulted in scrambling of the deuterium into the aryl positions before 0-insertion occurred, suggesting a possible benzyne intermediate in the reaction shown in eq 5.'4c (a) Tuan,D. F.-T.; Hoffmann, R. Inorg. Chem. 1985, 24, 871. (b) Bottomley, F.; Brooks, W. V. F. Ibid. 1977, 16, 501. (c) Vaughan, G. A,; Hillhouse, G. L.; Buchwald, S. L., manuscript in preparation. (15) (a) Lubben, T. V.; Wolczanski, P. T. J . Am. Chem. SOC.1985, 107, 701. (b) Lubben, T. V.; Wolczanski, P. T. Ibid. 1987, 109, 424. (c) Brindley, P. B.; Scotton, M. J. J . Chem. Soc., Perkin Trans. 2 1981, 419. (d) Blackburn, T. F.; Labinger, J. A,; Schwartz, J. Tetrahedron Lett. 1975, 3041. (16) (a) van Asselt, A.; Santarsiero, B. D.; Bercaw, J. E. J . A m . Chem. SOC.1986, 108, 8291. (b) Sharpless, K. B.; Woodard, S. S.; Finn, M. G. Pure Appl. Chem. 1983, 55, 1823. (c) Also, ref 18. (17) Shell process for propylene epoxidation: (a) Sheldon, R. A. In Aspects of Homogeneous Catalysis; Ugo, R., Ed.; Reidel: Dordrecht, 1981; Vol. 4, p 3. (b) Shell Oil Br. Patent 1249079, 1971. Shell Oil US. Patent 3 923 843, 1975. (18) Applications in organic synthesis: (a) Finn, M. G.; Sharpless, K. B. In Asymmetric Synthesis; Morrison, J. D., Ed.; Academic: New York, 1985; Vol. 5, p 247. (b) Sharpless, K. B.; Behrens, C. H.; Katsuki, T.;Lee, A. W. M.; Martin, V. S.; Takatani, M.; Viti, S. M.; Walker, F. J.; Woodard, S. S. Pure Appl. Chem. 1983, 55, 589. (c) Sharpless, K. B.; Verhoeven, T. R. Aldrichim. Acta 1979, 12, 63. ( 1 1) 4,5, and 6 can be prepared independently: 3
5; 5
+H20
0002-786318711509-5539$01.50/0
exploring the scope of the reactivity of nitrous oxide with other do organometallic complexes, and these studies will be the topic of future reports. Acknowledgment. Financial support from the National Science Foundation (Grant CHE-8520329) and the donors of the Petroleum Research Fund, administered by the American Chemical Society (17718-AC3), is appreciated. The N M R facilities were supported in part through the University of Chicago Cancer Center Grant (NIH-CA-14599). Supplementary Material Available: Tables of analytical, NMR, and IR data; synthetic and experimental details; and kinetic data (4 pages). Ordering information is given on any current masthead page.
The First Structural Characterization of a Dimeric Lithium Ketone Enolate-Lithium Diisopropylamide Complex Paul G. Williard* and Mark J. Hintze Department of Chemistry, Brown University Prouidence, Rhode Island 0291 2 Received February 27, 1987 While attempting to obtain information concerning the effects of chelation on the aggregation state of lithium ketone enolates, we discovered a new dimeric, lithium enolate-lithium amide base complex. We obtained the structure of this dimeric complex by X-ray diffraction analysis. This dimeric complex may occur commonly in solution; hence, the general structure of the dimer is likely to have mechanistic implications in the reaction of ketones with lithium amide bases. When a heptane solution of the ketone 1 is added to a suspension of freshly prepared lithium diisopropylamide (LDA) in the same solvent, a homogenous solution is formed after only 0.5 equiv of the ketone has been added. If the addition of ketone to the base is stopped after the homogenous solution is obtained and if the reaction mixture is subsequently placed in a freezer maintained at -20 "C, large transparent crystals are obtained. In an alternative procedure, we added a full stoichiometric amount of ketone to the amide base, followed by addition of a second full equivalent of n-BuLi. This procedure also yields crystals of the same composition. We carried out low-temperature (- -100 "C) diffraction analysis of these crystals according to our standard protocol.] The structure of a dimeric, enolate-LDA complex was obtained.2 This is depicted in Figure 1 as the aggregate 2. Two computer generated plots of the dimer are shown in Figure 1 in order to emphasize its stereochemistry. The perspective of the molecule differs by 90" rotation about a horizontal axis in the two plots. In the top plot, the seven-membered chelate rings, made up of an enolate oxygen, a lithium atom, and the silyl ether oxygen, are clearly discernible. In the bottom plot, the slight curvature of the fused, four-membered ring skeleton of the dimer is apparent. It is noteworthy that the silyl ether oxygens coordinate with lithium atoms despite the steric bulk of the tert-butyldimethylsilyl group. This coordination is contrary to recent observations that silyl ethers are poor electron donor^.^ ~~~~
( I ) Williard, P. G.; Carpenter, G. B. J . Am. Chem. SOC.1986, 108, 462.
(2) The enolate-LDA complex underwent spontaneous resolution and crystallized in the noncentrosymmetric, monoclinic s ace group P2, with unit cell parameters a = 12.282 (3) A, b = 10.91 1 (3) c = 17.606 (8) A, and p = 92.67 (3)'. The unit cell contained two asymmetric units of molecular formula [(C13H27Si02Li).(C6H,4NLi)]2 in a volume of 2356.8 (1.3) A3. This produces a calculated density of 1.01 gem-'. A total of 4100 reflections were recorded by using the 8:28 scan routine and graphite monochromated Mo Kcu radiation in the range 3.5O 5 28 5 45O. The final agreement factors are R = 0.0414 and R , = 0.0493 for 450 parameters and 3557 unique, observed reflections. A complete description of the crystallographic parameters is submitted as Supplementary Material.
{
0 1987 American Chemical Society
